How to graph probability of death by a continuous variable in Stata?

[Eugene DePasquale]

Hi,

Can anyone help me with how to graph probablity of death (at 1 year) by a continuous variable (such as age) in Stata? For an example of what I'm trying to do see the attached picture.

Thanks!

[Joseph Coveney]

You don't give any information about what you're starting with, but I suppose you could try something along these lines.

sysuse auto
logit foreign c.trunk, nolog
summarize trunk, meanonly
local numlist `=r(min)'(`=(r(max)-r(min))/50')`=r(max)'
quietly margins, at(trunk=(`numlist'))
marginsplot , plotopts(msymbol(none))

You'll probably want to tweak things a little to get it to look like you want, for example, rendering of the confidence interval band (I assume that that's what you're showing in your picture) by marginsplot.

[Eugene DePasquale]

Thank you for your help. I am trying to look at probability of death following transplant with continuous variable such as age or creatinine.
It looks like this works with stcox.. Is there a way to change the y-axis to be probability of death? Also how do I change the scale of the x-axis (options in graph edit did not let me move past the e notation)?

stcox ischtime, nolog
summarize ischtime, meanonly
local numlist `=r(min)'(`=(r(max)-r(min))/50')`=r(max)'
quietly margins, at(ischtime=(`numlist'))
marginsplot , plotopts(msymbol(none))

Let me know if this coding looks ok.

Thanks,
Eugene